© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130982
    Prev
    Next
    INDEX
    Explanations

    The neuron fires most strongly on numeric tokens (i.e. numbers or number‐like strings) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    code punctuation and numbers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    や
    -1.06
    以及
    -1.05
     illetve
    -1.01
     وكذلك
    -0.94
    还有
    -0.86
     ebenso
    -0.85
     electrophoresis
    -0.84
    そして
    -0.82
     многие
    -0.82
     ¨
    -0.81
    POSITIVE LOGITS
    cierre
    0.91
     bordado
    0.87
    zzlement
    0.86
     nepri
    0.82
     cvi
    0.82
    shout
    0.79
     índ
    0.79
     blusa
    0.79
     ativa
    0.78
     sapat
    0.78
    Activations Density 0.057%

    No Known Activations